Daniel Ruiz, Luca Luceri, Emilio Ferrara · 2026 · arXiv
DOI: 10.48550/arXiv.2610.30418 · open-access PDF
Dieser Mosaik-Effekt schafft ein kritisches Risiko für Agenten, die über vertrauliche Daten operieren. Wir quantifizieren seine Realisierung mit 30 synthetischen Beispielen für Cybersicherheit, Medizin und Militär, die 10.800 endgültige Antworten von drei LLM-Agenten in kontrollierter Simulation und 1.440 Antworten von Live-Agenten in zwei Open-Source-Anwendungen liefern.
Machine translation (Meta M2M100); the English original follows.
Modern LLM agents can retrieve information from accessible documents, and then disclose a sensitive conclusion assembled from them. This mosaic effect creates a critical risk for agents operating over confidential data. We quantify its realization with 30 synthetic cybersecurity, medical, and military examples, yielding 10,800 final answers from three LLM agents in controlled simulation and 1,440 answers from live agents in two open-source harnesses. Even under benign queries designed to be answerable from one document, GPT-5.6-Sol, the best performing model tested, discloses forbidden composi
Read in: English · Spanish · Chinese · Hindi · Arabic · Portuguese · French · German · Japanese · Russian · Korean · Indonesian · Vietnamese
GET https://store.agentexchange.work/gw/papers/verify?title=...&year=... (0.003 USD), fetch this record as JSON with /gw/papers/paper?id=2610.30418, or list what it cites with /gw/papers/references?id=.... Pay per call in USDC (x402) or with a 19 USD card key for 1,000 calls. Catalogue · llms.txtSources: OpenAlex (CC0), arXiv, Crossref. Agent Exchange